Skip to main content

No project description provided

Project description

Alectio-CLI :rocket: :rocket:

Alectio-cli is Alectio's's official command-line interface (CLI)wrapper. It allows you to create projects, experiments and do hybrid labeling tasks.

Configuration

Setup

These setup instructions are for CLI usage only.

We recommend installing alectiocli in your virtualenv rather than doing a global installation so you don't run into unexpected behavior with mismatching versions.

pip insall alectiocli

Authentication

For usage of cli, you first need to authenticate with Alectio's server.

alectio-cli --login

alt text This will redirect you to Alectio's platform and once you login, it will generate a authentication token, which you need to copy and paste it on the command line.

If you wish to force login or refresh token after the authentication is done, use:

alectio-cli --login --refresh

Usage

If you wish to get more information about any command within alectio-cli, you can execute alectio-cli --help command.

Common alectio-cli --help Options

  • login : Login for token authentication
  • project : A sub-app which handles all project related tasks
  • experiment : A sub-app which handles all experiment related tasks

Common alectio-cli project --help Options

  • create : Create an alectio project
  • list : List all the projects of the user
  • delete : Delete a project
  • hybrid-labeling:

Projection Creation:

alectio-cli project create [OPTIONS] YAML_FILE LABEL_FILE

A folder Alectio_cli_sample_yamls will be generated which will contain sample YAML files for doing various tasks like:

  • Yaml file for creating an experiment
  • Yaml file for manual curation experiment
  • Yaml file for Hybrid Labeling Task

Sample YAML FILE for creating project

name: "Sample Project"
description: "test "
type: "Image Classification"
alectio-dataset: "false"
test_len: 0
s3_bucket: 
data_name: 
alectio_dir: 
pre_loaded_model: 
project_type: "curation"
data_format: "image"
docker_url: 
premise: "true"
train_len: 10000
dataset_source: "on-prem"
labels: True

List Projects:

alectio-cli project list

Project Deletion:

alectio-cli project delete <proj_id>

Hybrid Labeling Task:

alectio-cli project hybrid-labeling [OPTIONS] YAML_FILE

Experiment

Common alectio-cli experiment --help Options

  • create : Create an alectio experiment
  • run : Runs an alectio experiment

Experiment Creation

alectio-cli experiment create [OPTIONS] YAML_FILE

Sample YAML FILE for creating project

project_id: '22977034a12b11ecaf91cbe75e9b38c0'
name: 'Test Experiment'
n_records: 100
limits: False
qs: [] #Empty for auto_al,ml_driven. To be filled for manual curation.
seed: 120
labeling_type: pre_labeled
is_sdk_setup: False
is_curr_fully_labeled: False
assoc_labeling_task_id: ''

Run Experiment

alectio-cli experiment run [OPTIONS] PYTHON_FILE EXPERIMENT_FILE

ℹ️ Note that this cli is still in development phase, instability might occur.

Future

We are continually expanding and improving the offerings of this application. Some interactions may change over time, but we will do our best to retain backwards compatibility.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

AlectioCLI-0.1.3.tar.gz (14.3 kB view hashes)

Uploaded Source

Built Distribution

AlectioCLI-0.1.3-py3-none-any.whl (18.7 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page